Profile

Duncan Murray profile image

Duncan Murray

Student

  • 25 reviews
  • 24 completed
Content 
Instructor 
Provider 
Almost finished this, but so far this is the best course I have taken - it is really well done and Andrew is a great teacher. You end up with practical skills on machine learning, and although the maths looks quite complex he takes the time to explain it well. This is also the only MOOC I have come across which has: \- complete wiki pages, well laid out with all key facts \- in video quizzes which are actually timed correctly (why do some courses have questions about things they haven't talked about yet?) \- well designed assignments. I like the template idea, as that is not unlike what happens in the real world \- great feedback system on the assignments (at each stage they build in tests that you can check to see how you are going before you have to submit the assignment) I would absolutely recommend this course to anyone interested in the subject, as although there is a lot of maths, you will end up with practical skills at the end (and Octave is pretty impressive tool to learn)
Content 
Instructor 
Provider 
I don't think this coursera offering is done in the spirit of the other online courses - the information page mentions that if you want course credit you can 'buy now' and spend $$$ to get a university credit. Personally I am very interested in information security, but the content was too light and aimed at a CEO level - seems to be more focused on the risk management plans than anything else (which not many people would really get to implement). The lectures consisted of seemingly endless guest speakers, some interesting, but overall too many lectures for the amount of information learned. I would not recommend this course, unless you were actually going to be writing risk management plans - the information security content could easily be learned elsewhere without listening to many hours of lectures. On top of all that - they changed the fact that you would get any certificate at all. Originally they said you'd get a certificate of participation but this was changed early on that you wouldn't unless you paid the money to get the full university credit. It is currently being discussed on the course forums, and I would hope that they provide *something* saying I did the course and achieved XX% on the quizzes, as it was quite a slog to get through this course.
Content 
Instructor 
Provider 
If you want to build websites you should take this course - it has a high focus on user testing, which isn't always what a lot of developers like to do. One of those necessary but valuable chores, and in the course you will need to get real users to test your interface and you even need to provide photos of them doing testing it. It felt a bit weird asking friends to get photographed testing a 'dummy' web application but it is one of things you need to do. Scott is a good speaker, and knows his stuff, but you won't learn any web coding techniques as it is focused on the User Interface, so if you just want the web building skills you may want to try the Udacity course (build a web site / blog ?) The peer review system in this course is very good - every student must pass an evaluation test to make sure they mark other students assignments correctly (based on the marks given by TA's) - this worked well. You are provided with free tools to build the prototypes which have an interactive feel to them - they are nice but a bit clumsy and very slow, so I chose to build a web site and host it - this took a lot more time, than I estimated in terms of the rest of the course. My only gripe with this course is that the certificate at the end doesn't really reflect the amount of effort you put into this course. A friend also did the course without doing any assignments and the only difference between our certificates is the word 'distinction' on mine in a blue box on the PDF. I would have preferred to see the grade on there, showing that I had done all the assignments well (though perhaps coursera is planning on having this available as a service later) Overall - a very worthwhile course.
Content 
Instructor 
Provider 
Highly recommend this course to anyone interested in design (of any field). The course is quite practical and you will need to work to complete the homework and assignments, but it gives good techniques on the design process. Karl is an interesting lecturer and it is also a lot of fun - you get to practice drawing [chairs], and get to design whatever you like as a final project. Personally, I would have like a little bit more on specific techniques on designing 'well' - what colours match, ergonomics - but realise this is much more than can be achieve in a single course (would love to see a design 2 course)
Content 
Instructor 
Provider 
I am an experienced programmer and have recently learned Python so I took this course to test myself and remove any bad habits I had picked up. It was a really useful, and lots of fun - the web version of the python environment works beautifully, lectures are clear and interesting and the quizzes are spot on. The lecturers are engaging and the assignments are the best part - writing a series of short games, each increasing in complexity just enough to test new concepts. Great course - looking forward to the next Python course done by the same people!
Content 
Instructor 
Provider 
Seeing people in the Data Science stream will have already done R programming (which covers the installation of R - Studio), then the bulk of this course is a couple of lectures with the final "Assignment" writing a one line MD file, then clicking "Fork" on github. That is a terrible excuse to charge people for this course, when the content should be part of another sections introduction.
Content 
Instructor 
Provider 
I come from a computing background so was no stranger to Logic as such, but took this course to understand the formal language and syntax I have seen used. Mission Accomplished - they did a great job teaching this. The videos were interesting, clear and useful. There plenty of course notes and course wiki note summaries. It was hard work though - most of the quizes involved a fair bit of work, like writing out proof trees. There is a peer graded exam, so you'll need to hope the people marking you understand the topic as well you do. My only suggestion for change would be to keep the peer graded assignments, but change the exam to a more traditional one (autograded), but overall - a great course and I learned a lot from it.
Content 
Instructor 
Provider 
I had no experience with Game Theory, but some exposure to AI and software development in general. The lectures are a bit too short and basically focus on the slides. The bulk of the course is around you setting up the pre-built Stanford Game playing software and implemented various game playing techniques through it. I found the mechanics of using this software a useful exercise, but would have liked to implement something myself instead (or as well as)
Content 
Instructor 
Provider 
My background is in programming with some additional Math classes, and I took this course to get a better understanding of the theory and logic behind CS. The lecturer is easy to listen to and explains concepts well, though it is mostly reading from the slides (though with this type of material I cant think of a better way to do this). The information is very theoretical, though it gets interesting when you apply it to known concepts such as regular expressions and SQL like operations
Content 
Instructor 
Provider 
I thought I had a lot of experience with Metadata, but this course gave me a new level of understanding around the types of schemas used. The lecturer is very good, participated in the forums and even included a topical video on phone metadata when the US news story appeared on logging phone metadata.
Content 
Instructor 
Provider 
I have a programming background so found this course really enjoyable - the 4 week burst of learning R was spot on in my opinion. If you are new to programming you may find it a bit rushed, but still should be able to complete it.
Content 
Instructor 
Provider 
I had some prior experience with this (Model Thinking course), but still got useful information from this course. It is a great, compact course giving a good introduction to Networks. The lectures are interesting, but the best part is that all videos and quizzes are available from day 1, so if you have done some of this before, you can move through the course quickly at your own pace. I wish more courses would offer this method.
Content 
Instructor 
Provider 
Good, frantic paced introduction to getting a startup going, but I think it should be either longer or split into 2 courses, as there are 2 main areas: \- Technical side is great and goes into the detailed tasks of building a site \- Business side - which covers (some of) the marketing, funding. I completed the technical assignments but did not have a project ready to market and try and get investors.
Content 
Instructor 
Provider 
I found this course to be quite hard, although it was well explained. You do useful implementations of Linear programming, such as an optimal Diet planner.
Content 
Instructor 
Provider 
This is a brilliant course - short but you get a lot out of it. I hadn't done any serious maths for many years and was taking Calculus 1 at the same time, so don't be put off by the 'Two' thinking you need a calculus background - it was very self contained, and Jim Fowler really is a great maths teacher.
Content 
Instructor 
Provider 
This is an excellent Calculus course - I have never had a maths class where the Professors made the content so interesting and with such enthusiasm. They also produced a free ebook as part of the package (Mooculus) which was quite helpful to practice on additional examples.
Content 
Instructor 
Provider 
This course is a pretty good attempt at combining Maths theory with practical programming - I learned a lot about Python comprehensions, and a fair bit about Linear Algebra. It gives a background on the maths and appears to finish where the Machine Learning course begins - though I felt I could apply the learning from ML theory a lot easier than from this course. PROS \- great boot camp which forces you to really write correct Python code \- learning the Maths theory and applying it in code each week \- grader allows multiple attempts with no obvious time limits (necessary though) \- active forums \- fun secondary (still mandatory) assignments - I loved the one on changing the perspective of a photo CONS \- Overly picky autograder, often rejects correct Python code because it needs to be written in a certain way (possibly because they force you to use comprehensions in a certain way) \- lots of homework: 2-3 short programming assignments every week \- lectures could use some more comments around applications of the theory being used, rather than just the maths theory followed by occasional code. The secondary weekly assignments should have been referred to frequently in the videos, to reinforce why this particular bit was useful Overall, I am glad I did this course - it was very challenging and time consuming but well worth the effort.
Content 
Instructor 
Provider 
This course was so short that I didn't really feel like I was in doing it - but I am glad I did it. The reverse engineering exercise seemed daunting at first, as I had never done serious assembly programming before, but it was fun to do (took about an hour). There were a lot of general facts discussed, and the quiz's basically tested that you watched the videos, though he did a good job of giving an interesting overview of the malware situation. You definitely need to be a programmer to pass this though - I used Python, but the exercises are calculation style so you can use any language. Since it was such a short course you won't get into any *real* depth of cracking the malware, but I would recommend this course to get a current overview and want to play with some assembler.
Content 
Instructor 
Provider 
This is an excellent course, but quite difficult unless you have a strong maths background and good programming skills. Interesting, well planned lectures, fun assignments but the quizs seemed pretty harsh - get too many answers wrong and it says "don't cheat!". The best takeaway from this course was the confirmation to me that you should NOT implement your own crypt functions unless you really know what you are doing and test it mathematically.
Content 
Instructor 
Provider 
I really wanted this course to be great - it had a very wide range of topics, they supply a VM with all the software (had some issues though), and the lecturers had a lot of energy to make it interesting. The programming assignments I did were simple, but still interesting and great for 'wow factor' to people who hadn't done this before Reason I dropped this course was: a) could not get the VM to work with one library, after several of hours fault finding. This was needed to get one of the assignments done. b) was not really interested in some of the assignments (a bit too diverse - going from programming to reporting) c) I had 3 other courses on at the same time, so didn't really have the time to commit. I will retake this course the next time it is offered, but hopefully a couple of the topics could be trimmed, or maybe split into 2 courses (programming focus another on the BI focus)
Content 
Instructor 
Provider 
Very interesting topic and they do a reasonably good job of making interesting what is quite a maths/logic heavy course. I would recommend this if you are interested in AI, and it is great that they offer the multiple levels of completion - I like the idea of a certificate to show what level you 'worked' on the course. I completed this at an awareness level only (videos and quizs), but will be retaking it in Jan 2014 to do the assignments [the first time I did this I didn't have the time to do the programming assignments] To improve this course, it needs a better set of wiki pages containing the key topics from the videos.
Content 
Instructor 
Provider 
Overall he does a good job of giving broad overview of a fairly diverse subject. Others have complained it was too shallow, and perhaps some parts were a bit glossed over, but to cover them all would make it a 50+ week MOOC. I have a good programming and SQL background, but still found the assignments to be difficult, seeing there was no test data or feedback which made it unclear. I would recommend this course if you wanted a good overview or the topics, but the assignments and exams offer no feedback, so if you are aiming for a certificate you would need to be prepared to work at it. PROS \- gives a taste of most of the topics \- speaks slowly in the starting lectures to allow people to get used to his accent (I had no trouble understanding, even on a smartphone) \- they provide the lecture slides in a single ZIP file to download. Very useful, but not as effective as some wiki pages CONS \- no useful wiki page. I take notes but found it difficult to find where the information was for certain things - kept having to go through the videos \- the assignments need rethinking. They are kind of 'simple' once you *finish* them, but it is quite difficult to know if you are on the right track. You only get a couple of shots and entering a value in the answer but have no way of knowing if you are on the right track. A better approach would have been to give some test data which shows what is expected of that. \- I found the exam was difficult seeing it was so hard to find which video the information was buried. I would advise taking notes as you go. \- Just found out the grading will be on a sliding scale. This is a terrible idea, I would rather know how *I* did against the topic instead of being compared to a heap of other people I dont know.
Content 
Instructor 
Provider 
This is quite a hands on / practical course - it gives you the tools and practice to build some programs to analyse the stock market. The lecturer seems to know what he is talking about, and comes across as a business teacher rather than an academic - not a bad thing in itself, but sometimes there are gaps in the explanations. If you are interested in getting started in this field, I would recommend this course, but be prepared for some extra work. I have a good programming background and found the assignments quite hard. This may have been my lack of experience with python / stock markets but I spent a lot of time trying to work out what seemed like basic coding issues (like, what the hell does this line *do* ) Having said that, I had no trouble setting up the environment and libraries - the first assignment takes you through this and validates that you have it all installed. PROS \- interesting topic \- you write a program to actually analyse the market CONS \- audio quality of lectures are too low to hear effectively on a smartphone \- later assignments not too clearly explained \- not a fan of the music at the start of every lecture (annoying when you watch a series all at once)
Content 
Instructor 
Provider 
This is a fun, easy course to do and worthwhile if you are interested in what to do and what not to do if you are considering gamifying a business process / website. The homework quizs are multiple choice but unlike other MOOC's you lose 20% each time you take a quiz so think about the answers first. The peer reviewed assignment is basically building up the first stages of a business case and doesn't take a lot of time to finish.
Content 
Instructor 
Provider 
A very interesting introduction to various statistical models - the lecturer explains things well and sounds like Alan Alda from MASH which is a cool bonus. I had a basic maths background, and some of the maths look hard but turned out to be fairly basic (well explained). The quizs offered good feedback, and there are no peer assignments or programming homework. The exam was practical in that it tested that you could do some of the calculations so it isn't just multiple choice theory, so you'll want to make sure you do follow the quizs.